We are interested in the identification and analysis of genes which increase susceptibility to cancer and are taking a three-pronged approach to addressing the problem. We have used linkage analysis to look for highly-penetrant Plostate cancer susceptibility loci by performing a ganomc-wide scan in 266 high-riskI prostate cancer families, preliminary results from this scan implicate multiple lociof interest, and ongoing efforts focus on fine mapping of the regions identified and positional cloning of the affected genes. We have used population-based case/control studies to determine the role played in sporadic breast cancer of mutations in the highly-penetrant genes BRCA1 and BRCA2. Work to date has focused on a population of women in Western Washington State, and ongoing efforts center around 2 much larger studies, 1 in the U.S. and 1 in Shanghai, China, and also on understanding the effects of non-genetic factors on panetrance of BRCA1 and 2 mutations. Our human work has underscored the difficulties of studying a genetically complex disease in human families. We have therefore developed the domestic dog as a uniquely powerful system for advancing cancer genetics, taking advantage of both the large family size as well as the closed breeding populations (breeds) that characterize purebred dogs. To date, we have developed a map of the dog genome and demonstrated the utility of the canine system for simplifying the locus heterogeneity associated with mapping complex traits. In the process we have mapped several disease genes, including a kidney cancer locus, of interest to the human genetics community. The next key step is to integrate our throe approaches. Cloning prostate cancer susceptibility loci identified in high risk families and analysis of mutation frequency and distribution will allow us to determine the significance of these genes in the general population, much as we have done in our studies of breast cancer. Mapping and cloning of canine cancer loci will accelerate identification both of highly penetrant genes and of less penerrant genetic modifier loci, and will permit us to test the hypothesis that genes we identify in dogs are equally relevant in human disease, as suggested by a variety of prior studies. This K05 will provide the resources, and particularly the time, necessary to provide guidance and mentoring to students and postdocs in my lab as they pursue the new avenues of research that will achieve the integration of our efforts.